browser/components/translation/cld2/internal/fixunicodevalue.cc@6474c204b198 (annotated)
browser/components/translation/cld2/internal/fixunicodevalue.cc
Wed, 31 Dec 2014 06:09:35 +0100
- author
- Michael Schloh von Bennewitz <michael@schloh.com>
- date
- Wed, 31 Dec 2014 06:09:35 +0100
- changeset 0
- 6474c204b198
- permissions
- -rw-r--r--
Cloned upstream origin tor-browser at tor-browser-31.3.0esr-4.5-1-build1
revision ID fc1c9ff7c1b2defdbc039f12214767608f46423f for hacking purpose.
| michael@0 | 1 | // Copyright 2013 Google Inc. All Rights Reserved. |
| michael@0 | 2 | // |
| michael@0 | 3 | // Licensed under the Apache License, Version 2.0 (the "License"); |
| michael@0 | 4 | // you may not use this file except in compliance with the License. |
| michael@0 | 5 | // You may obtain a copy of the License at |
| michael@0 | 6 | // |
| michael@0 | 7 | // http://www.apache.org/licenses/LICENSE-2.0 |
| michael@0 | 8 | // |
| michael@0 | 9 | // Unless required by applicable law or agreed to in writing, software |
| michael@0 | 10 | // distributed under the License is distributed on an "AS IS" BASIS, |
| michael@0 | 11 | // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| michael@0 | 12 | // See the License for the specific language governing permissions and |
| michael@0 | 13 | // limitations under the License. |
| michael@0 | 14 | |
| michael@0 | 15 | // |
| michael@0 | 16 | // Routine that maps a Unicode code point to an interchange-valid one |
| michael@0 | 17 | // |
| michael@0 | 18 | |
| michael@0 | 19 | #include "fixunicodevalue.h" |
| michael@0 | 20 | #include "integral_types.h" |
| michael@0 | 21 | |
| michael@0 | 22 | namespace CLD2 { |
| michael@0 | 23 | |
| michael@0 | 24 | // Guarantees that the resulting output value is interchange valid |
| michael@0 | 25 | // 00-FF; map to spaces or MS CP1252 |
| michael@0 | 26 | // D800-DFFF; surrogates |
| michael@0 | 27 | // FDD0-FDEF; non-characters |
| michael@0 | 28 | // xxFFFE-xxFFFF; non-characters |
| michael@0 | 29 | char32 FixUnicodeValue(char32 uv) { |
| michael@0 | 30 | uint32 uuv = static_cast<uint32>(uv); |
| michael@0 | 31 | if (uuv < 0x0100) { |
| michael@0 | 32 | return kMapFullMicrosoft1252OrSpace[uuv]; |
| michael@0 | 33 | } |
| michael@0 | 34 | if (uuv < 0xD800) { |
| michael@0 | 35 | return uv; |
| michael@0 | 36 | } |
| michael@0 | 37 | if ((uuv & ~0x0F) == 0xFDD0) { // non-characters |
| michael@0 | 38 | return 0xFFFD; |
| michael@0 | 39 | } |
| michael@0 | 40 | if ((uuv & ~0x0F) == 0xFDE0) { // non-characters |
| michael@0 | 41 | return 0xFFFD; |
| michael@0 | 42 | } |
| michael@0 | 43 | if ((uuv & 0x00FFFE) == 0xFFFE) { // non-characters |
| michael@0 | 44 | return 0xFFFD; |
| michael@0 | 45 | } |
| michael@0 | 46 | if ((0xE000 <= uuv) && (uuv <= 0x10FFFF)) { |
| michael@0 | 47 | return uv; |
| michael@0 | 48 | } |
| michael@0 | 49 | // surrogates and negative and > 0x10FFFF all land here |
| michael@0 | 50 | return 0xFFFD; |
| michael@0 | 51 | } |
| michael@0 | 52 | |
| michael@0 | 53 | } // End namespace CLD2 |
| michael@0 | 54 |
